Errors of measurement of exposures and outcomes are frequently identified as an important source of uncertainty in cancer and respiratory epidemiology. Recently there has been a renewed interest in studying statistical methods for treating this problem. In part, this interest can be traced to the growing use of more sophisticated techniques for determining exposures and outcomes, as in a study conducted by our lab on the effect of maternal smoking and neonatal lung function in which exposures are in part derived from repeated measurements of urine cotinine using radioimmunoassay. There is also a growing awareness of inaccuracies in traditional instruments such as questionnaires. In the Nurses Health Study, we have studied measurement error in the major source of nutritional exposure data, the food frequency questionnaire, by conducting validation studies in which more accurate daily diet records are collected for a subset of women in the study. The long term objectives of this application are 1) to continue to develop statistical methods useful in practical research settings, 2) to apply these techniques in the context of scientifically important data analyses, and 3) to implement the methods in the form of distributable computer software. The specific topics to be addressed include: methods for survival analysis with errors in time-dependent covariates; pooling of regression coefficients from several studies with possibly different measurement error distributions; measurement error in both categorical and continuous independent variables in logistic regression; treatment of measurement error resulting from the use of bioassays in epidemiology; and, importantly, several topics relating to design issues involving measurement error, such as the optimal size of internal and external validation studies, and the development of two-stage sequential designs. The methods will be applied to data from important data sets in cancer and respiratory epidemiology, including the Nurses Health Study, the Health Professionals Follow-up Study, the Six Cities Study, and a prospective study of the health effects of maternal smoking on neonatal lung function. We believe that methods developed for these diverse studies will be of use in many epidemiologic settings.